Between weaker than expected electric vehicles sales (at Tesla (TSLA) in particular) and rapid expansion of wafer and chip capacity, it’s been a rough stretch for semiconductors leveraged to silicon carbide (or SiC) power semiconductors. With auto demand looking weaker, industrial end-markets still correcting, and capacity still coming online in China, the near-term outlook is treacherous at best.
Shares of onsemi (NASDAQ:ON) have had a rough go of it since my last update, with the shares down about 14%. That is dwarfed by the 70% rise in the SOX (though AI plays like Nvidia (NVDA) distort that comparison some), and other diversified SiC players like Infineon (OTCQX:IFNNY) and STMicroelectronics (STM) have likewise done better, though onsemi has at least outperformed Wolfspeed (WOLF) and Rohm (OTCPK:ROHCY).
While I am concerned about the risk of a multiyear SiC glut, and onsemi is less diversified than STMicro or Infineon, I do think this is a time to consider the shares. Management has made fundamental improvements to the business that have preserved margins in this downturn, and I still believe in the long-term value of onsemi’s power and sensor franchise, as well as its leverage to long-term electrification growth across a range of end-markets.
A Modest Beat Seems Likely, But What About Guidance?
Management at onsemi has pretty much dialed in the guidance game, as the company has a track record stretching back through 2022 of modest single-digit quarterly revenue beats coupled with modest (<50bp) gross margin beats.
Given that, I don’t think the actual reported results from the upcoming second quarter earnings will be hugely significant, as the company will probably again post a modest beat but stay within management’s guidance; in this case, that likely means revenue around $1,750-$1,760M, or a roughly 16%-17% year-over-year drop, with gross margin around 45.3% or 45.4% (versus 47.4% last year and 45.9% last quarter).
It’s the guidance that’s going to be the big driver, and particularly whether the company sees signs of improving demand that would support sequential revenue growth in the second half of the year in the auto and industrial businesses.
I’m skeptical, but not because of any shortcomings with onsemi or its management. Rather, it’s a customer mix and industry growth concern. EVs are still growing as a category and onsemi customer Geely (OTCPK:GELYY) has been looking stronger, with 117% yoy growth for its “new energy vehicles” in the first half of 2024. Elsewhere, Volkswagen and BMW likewise appear to be picking up a little bit, and onsemi should have new customer/model launches to further help auto sales recover in the second half.
Countering that, Tesla (TSLA) is not looking particularly strong (Q2 deliveries were down 5% yoy). Tesla is a disproportionately large share of the auto SiC market (around 75% of 2023 SiC demand, according to Yole Group, and likely 50%+ in 2024), and I’m concerned about how that all nets out for onsemi.
Likewise, providers of industrial automation and electrification products (Cognex (CGNX), Rockwell (ROK), et al) continue to report sluggish conditions, so I’m not expecting much from the industrial end-markets until maybe Q4 when there should be more inventory restocking.
Is A Lasting Glut Of SiC On The Way?
I’ve seen radically different estimates of SiC wafer capacity among various third-party research providers, but all agree that a large amount of capacity was added in 2023 (50% to 100% yoy growth depending on the source) and a lot is still coming online in 2024. With EV sales falling short of expectations in Western markets and Tesla in particular losing share, I do have concerns that there won’t be enough demand in 2024/2025.
Making matters potentially worse, Chinese SiC production is ramping up. Local players like SICC and TankeBlue are adding capacity and BYD (OTCPK:BYDDY) is at least dipping its toes into internal SiC MOSFET production. This raises the risk of the Chinese EV market, the strongest EV market so far, turning away from Western suppliers; while I see less risk to STMicro, Infineon, and onsemi relative to Wolfspeed or Rohm, it’s still not a positive development.
It’s also important to note that auto demand really dominates the SiC market today. About 70% of SiC volumes go to auto customers, with around 12% to 13% going to industrial applications and a similar amount going to energy (renewable energy in particular). I am bullish on the growth potential of SiC within both of these markets, but I don’t see them growing fast enough soon enough to resolve this potential issue.
Last and not least, while it’s not a winner-takes-all market, somebody is going to lose. In the recent past, onsemi has projected future SiC market share of 35% to 40%, while STMicro, Infineon, and Rohm have all projected 30%. I’m no math scholar, but that math doesn’t work, particularly when you include the likes of Wolfspeed and a host of Chinese manufacturers. Based on strong customer relationships and product designs I think onsemi will be among the winners, but I think the point stands that there is going to be a reckoning and shaking-out process.
There is some good news, though. STMicro and Infineon have both revised their SiC forecasts lower and Wolfspeed has frozen its plans for a new facility in Germany. For its part, onsemi’s capex guidance suggests lower spending on SiC capacity as well, so the industry does seem to be behaving rationally with respect to capacity – something that hasn’t traditionally been true of semiconductor companies.
The Outlook
I don’t really take issue with onsemi management’s projections of double-digit market growth over the coming years, as I do think widespread electrification – building and factory automation, mobility, renewable energy, et al – will drive very strong demand for advanced power semiconductors (both SiC and GaN varieties).
I think that between energy infrastructure, factory automation, EVs and so on, onsemi will continue to see strong growth. I likewise am bullish on its sensor business; Sony (OTCPK:SNEJF) and Will Semi are threats to watch in image sensors, but I believe onsemi can ride the wave of increasingly sophisticated ADAS systems, as well as growing use of in-cabin imaging (for driver and passenger monitoring). Beyond that, I expect the company’s sensors to play a meaningful role in machine vision, including robotics and edge monitoring devices.
Relative to my last update, onsemi didn’t perform too badly in terms of financial performance (a 2% revenue shortfall and a 40bp OPM shortfall), but the auto and industrial markets have weakened considerably since then. With that, my 2025 revenue estimate is more than 20% below my prior number. I do think some of this can be “recaptured” later, but my 2033 revenue estimate is still about 14% lower than before and my long-term revenue growth estimate goes to about 7.5% (from 8%). I am aware of the risk of overcorrecting, and as I said, I’m still a believer in onsemi’s long-term leverage to power chip demand, but I think caution is prudent for now.
For margins, I’m very impressed that onsemi may bottom out at a gross margin of 45% or slightly above – it wasn’t that long ago that onsemi couldn’t hit that level in good years. While my operating margin estimate has come way down (from around 33% to 28%), I do see operating margins recovering back into the low-to-mid-30%’s over time with stronger auto and industrial end-markets. For adjusted free cash flow, I still expect long-term FCF margins in the mid-20%’s, driving double-digit FCF growth.
Between discounted cash flow and margin-driven EV/revenue and EV/EBITDA approaches, I believe onsemi shares are undervalued. Long-term discounted cash flow gives me a fair value in the mid-$80’s, while those other approaches get me to the mid/high-$70’s. Keep in mind, though, that those only look ahead about a year, so they don’t really capture the recovery I expect. I could drag those estimates out into the future, and doing so would support a fair value in the $90’s, but at some point that devolves into a “choose your number and work backwards” exercise.
The Bottom Line
I can’t rule out poor guidance for the third quarter, weaker overall results in the second half of 2024, and/or a glut in the SiC market that takes more than a year or two to resolve. That’s just a fundamental risk with this company and stock. I do think, though, that management has done a very good job of controlling what it can, and I believe the company is well-placed to be a leader in the power and sensor markets for years to come. With that, I think this is a name to consider again.
Editor’s Note: This article discusses one or more securities that do not trade on a major U.S. exchange. Please be aware of the risks associated with these stocks.
Read the full article here
